Primary infection with human cytomegalovirus (CMV) during early pregnancy can be both personally and economically shattering for many American women. Congenital CMV infection of the fetus following exposure of immune naive mothers leads to deafness and neurological sequelae in up to 8,000 children/year in the US, making development of a CMV vaccine a top priority of the National Academy of Sciences. However current methods of CMV vaccine research are unable to reconcile the need for in vivo testing of vaccines/hypotheses with the severe species specificity of CMV. We propose to develop vaccine strains of CMV which are altered in key loci involved in immunomodulation and dissemination. In this way we propose to achieve the balance between immunogenicity and attenuation required of an effective vaccine. Importantly these vaccines will be constructed in rhesus CMV in Phase I and our hypotheses evaluated in the most relevant in vivo model of CMV pathogenesis known, the rhesus monkey, during Phase II. Successful completion of Phase II evaluation will lead to transfer of the technologies to human strains, clinical trials and commercialization of a universal juvenile vaccine. This methodology abrogates the need to trial live vaccines with unknown characteristics in human patients. PROPOSED COMMERCIAL APPLICATION: Universal Vaccination of young women against CMV is a Level I priority for the National Academy of Sciences. Studies suggest such vaccination to be highly cost effective making this vaccine a good target for an HMO funded, government mandated universal childhood vaccine.